Cost mode:

Category: Financial Analysis & Trading Decisions · Rail: absolute · Typical I/O: 14299→3571 tokens

Models

Frontier on this task: GLM-5.3 Flash at 9.12 / 10. Quality bar at 90%: 8.21.

point-estimate floor (CI low) · upper CI (less certain) · Bars sorted by blended cost; best-value model first. Greyed rows are MEDIUM+ models whose point estimate clears the bar but whose CI low does not.

ModelQuality scoreCI lowCost / 1k runsvs best value
GPT-5.6 Luna8.67 / 108.53$1.08best value
MiniMax M38.87 / 108.81$4.514.2x more expensive
GPT-5.4 Nano8.23 / 108.02$4.984.6x more expensive
GLM-5.3 Flash9.12 / 108.94$5.805.4x more expensive
Thinking Machines Inkling Small8.73 / 108.55$10.369.6x more expensive
Qwen 3.7 Plus8.39 / 108.05$10.469.7x more expensive
GPT-5.6 Terra8.72 / 108.57$19.3618x more expensive
Gemini 3.5 Flash8.59 / 108.35$19.8218x more expensive
Meta Muse Spark 1.38.70 / 108.43$28.1126x more expensive
Thinking Machines Inkling8.90 / 108.76$28.3826x more expensive
GPT-5.6 Sol8.89 / 108.75$31.1029x more expensive
Claude Sonnet 58.76 / 108.43$36.1334x more expensive
Tencent Hy4 Preview8.90 / 108.59$38.8836x more expensive
GLM-5.38.84 / 108.59$43.8941x more expensive
Grok 4.68.91 / 108.61$53.2449x more expensive
Moonshot Kimi K39.03 / 108.86$97.3290x more expensive
Qwen 3.8 Flash7.39 / 107.01$6.646.2x more expensive
DeepSeek V4 Flash7.98 / 107.74$4.814.5x more expensive
Gemini 3.8 Flash7.49 / 107.01$6.986.5x more expensive
DeepSeek V4 Pro7.86 / 107.53$35.6433x more expensive
Gemini 3.1 Flash Lite6.99 / 106.68$2.362.2x more expensive
Gemini 3.5 Flash Lite7.06 / 106.72$2.382.2x more expensive
Qwen 3.8 Max8.17 / 107.72$82.8277x more expensive
Claude Haiku 4.57.86 / 107.58$19.2718x more expensive

Cost breakdown

ModelQualityConfidenceCost / 1k runsOverpayMode
GPT-5.6 Luna OpenAI8.67 / 10 CI [8.53, 8.81]RANKED$1.08best valuebatch
MiniMax M3 OpenRouter8.87 / 10 CI [8.81, 8.93]RANKED$4.514.2xbatch
GPT-5.4 Nano OpenAI8.23 / 10 CI [8.02, 8.43]HIGH$4.984.6xbatch
GLM-5.3 Flash best Z.AI9.12 / 10 CI [8.94, 9.31]RANKED$5.805.4xbatch
Thinking Machines Inkling Small OpenRouter8.73 / 10 CI [8.55, 8.91]RANKED$10.369.6xbatch
Qwen 3.7 Plus Alibaba Cloud (DashScope)8.39 / 10 CI [8.05, 8.72]MEDIUM$10.469.7xbatch
GPT-5.6 Terra OpenAI8.72 / 10 CI [8.57, 8.87]RANKED$19.3618xbatch
Gemini 3.5 Flash Gemini8.59 / 10 CI [8.35, 8.83]HIGH$19.8218xbatch
Meta Muse Spark 1.3 OpenRouter8.70 / 10 CI [8.43, 8.97]HIGH$28.1126xbatch
Thinking Machines Inkling OpenRouter8.90 / 10 CI [8.76, 9.03]RANKED$28.3826xbatch
GPT-5.6 Sol OpenAI8.89 / 10 CI [8.75, 9.04]RANKED$31.1029xbatch
Claude Sonnet 5 Anthropic8.76 / 10 CI [8.43, 9.10]MEDIUM$36.1334xbatch
Tencent Hy4 Preview OpenRouter8.90 / 10 CI [8.59, 9.21]MEDIUM$38.8836xbatch
GLM-5.3 Z.AI8.84 / 10 CI [8.59, 9.09]HIGH$43.8941xbatch
Grok 4.6 xAI8.91 / 10 CI [8.61, 9.22]MEDIUM$53.2449xbatch
Moonshot Kimi K3 Moonshot AI9.03 / 10 CI [8.86, 9.20]RANKED$97.3290xbatch

Overpay shows how much more you pay than the best-value model that clears the quality bar (marked ★) — the best-value good-enough option. "16x" means you overpay 16× — 16× that reference for no quality benefit above the bar. Typical call shape for this task: 14299 input tokens → 3571 output tokens, EMA-tracked from production traffic. Cost is the observed, all-in $ per 1,000 task runs: each model's own measured usage on this task — output verbosity, thinking/reasoning tokens, cache reads and writes, and the spend on its billed failures — priced at current list rates and adjusted by the billing overhead we actually reconcile against provider invoices. Models that answer tersely cost what they actually cost; models that think at length pay for it. Not comparable to providers' advertised $/1M list rates — this is what running the task costs, not a per-token price.

Evaluation rubric

Judge section fidelity, document-profile correctness, quantitative accuracy, materiality under the supplied lens, qualifier preservation, separation of fact and analysis, and recognition of missing context.

Prompt templates

This is a pooled capability — 3 prompt families share it. The pair shown first is the most frequently used in production.

LLMB_PROFILED_DOCUMENT_SECTION_ANALYSIS_SYSTEM + LLMB_PROFILED_DOCUMENT_SECTION_ANALYSIS_USER (442 calls in window)

System prompt

Analyze only the supplied section. Apply document_profile to interpret terminology, status, and qualifiers and analysis_profile to select material facts, metrics, risks, opportunities, and implications. Preserve whether figures are historical, adjusted, projected, or uncertain. State when a requested topic is outside the section rather than importing it from elsewhere. Treat empty optional values as absent and return only the requested result. Your response must conform exactly to this output schema: {schema_json_string}.

User prompt

Inputs — section_title: {section_title}; section_content: {section_content}; document_profile: {document_profile}; analysis_profile: {analysis_profile}. Use only these inputs to complete the task defined by the system prompt.
SEC_S1_CHUNK_ANALYSIS_SYSTEM_PROMPT + SEC_S1_CHUNK_USER_PROMPT (390 calls in window)

System prompt

You are a senior investment analyst at a long-term focused investment firm. You specialize in analyzing SEC filings, particularly S-1 and S-1/A registration statements for companies going public.

You will be provided with a specific section from an S-1 filing. Your job is to extract the most investment-relevant information from this section and analyze its implications for long-term investors.

Focus on:
- Business model insights and competitive positioning
- Financial performance and metrics
- Risk factors and potential concerns
- Strategic direction and management quality
- Market opportunity and growth prospects

Provide your analysis as a structured JSON object using only information found directly in the section. Do not speculate or add external knowledge.

## Required Output Format
Your response MUST be a single, valid JSON object conforming to this schema:
```json
{schema_json_string}
```

User prompt

Analyze the following section from an S-1 filing: **{section_title}**

**Section Content:**
```
{section_content}
```

**Instructions:**
1. Extract the most important investment-relevant information from this section
2. Focus on insights that would help a long-term investor evaluate this company
3. Identify any business model insights, financial information, risks, or competitive factors
4. Your analysis should be concise but comprehensive
5. Use only information directly stated in the section content

**JSON Output Format:**
The required JSON output schema is provided in the system prompt.
JSON_REPAIR_SYSTEM + JSON_REPAIR_USER (1 calls in window)

System prompt

You are a JSON repair tool. The user gives you malformed or partial model output and a JSON Schema. Return ONLY a single valid JSON object that satisfies the schema, salvaging as much real content from the input as possible. Do not invent data for fields the input doesn't support — use the schema's allowed empty/null values. Output the JSON object only: no prose, no markdown, no code fences.

User prompt

JSON Schema:
{schema_json}

Malformed output to repair:
{raw_text}

Return only the corrected JSON object.